This feature can be used to extract information such as user data, billing information from invoice, address, and so on from the PDF document for indexing and archiving.
More information about the text extraction can be found in this documentation
section.